README.md

metadata

license: cc-by-nc-sa-3.0
task_categories:
  - question-answering
language:
  - cs

This dataset is converted from fewshot-goes-multilingual/cs_czech-named-entity-corpus_2.0 version of Czech Named Entity Corpus 2.0 (project info, link to data) using script convert_czner.py. For longer texts (>200 ws tokens), the script samples text around the selected entity. It always follows form "<initial 20 ws tokens>, ..., <sampled window>". Then it extracts category name for the entity, all occurences of such entity in the text, and creates simple json representation. For example:

{"label": "komerční instituce", 
"answers": ["Škoda Auto", "Unipetrolem", "ČEZ"], 
"text": "S loňskými tržbami přesahujícími 130,2 miliardy korun se Škoda Auto umístila před Unipetrolem , který měl tržby 74,5 miliardy korun , a ČEZ s tržbami 53,2 miliardy korun ."}

Dataset size:

train 9315
test 10878

Citation

In case you are using the dataset, please cite the original authors

 @misc{11858/00-097C-0000-0023-1B22-8,
 title = {Czech Named Entity Corpus 2.0},
 author = {{\v S}ev{\v c}{\'{\i}}kov{\'a}, Magda and {\v Z}abokrtsk{\'y}, Zden{\v e}k and Strakov{\'a}, Jana and Straka, Milan},
 url = {http://hdl.handle.net/11858/00-097C-0000-0023-1B22-8},
 note = {{LINDAT}/{CLARIAH}-{CZ} digital library at the Institute of Formal and Applied Linguistics ({{\'U}FAL}), Faculty of Mathematics and Physics, Charles University},
 copyright = {Attribution-{NonCommercial}-{ShareAlike} 3.0 Unported ({CC} {BY}-{NC}-{SA} 3.0)},
 year = {2014} }